This proposal addresses a fundamental issue in HIV biology-how CD4 T cells die during infection. The loss of this subset of lymphocytes underlies clinical progression to AIDS. Using cultures prepared with human tonsil and spleen tissue, we showed that >95% of dying CD4 T cells consists of resting bystander cells abortively infected with HIV. Reverse transcription is greatly slowed in these cells by the action of the SAMHD1 host restriction factor, causing incomplete transcripts of viral DNA to accumulate in the cytosol. These transcripts are detected by IFI16, a host DNA sensor, which launches an innate immune response by inducing inflammasome assembly, caspase-1 activation, cleavage of pro-IL-1, and pyroptosis. Pyroptosis is a highly inflammatory form of programmed cell death in which plasma membrane pores form and cytosolic contents, including proinflammatory cytokines, are released. Thus, during HIV infection, most CD4 T cells in lymphoid tissues appear to die by cellular suicide and not by a direct toxic effect of the virus. Pyroptosis also unites two pathogenic signatures of HIV infection-CD4 T-cell death and chronic inflammation-in a single process. We hypothesize that pyroptosis drives progression of HIV disease and that it persists, albeit at lower levels, in subjects receiving combination antiretroviral therapy (cART) and contributes to the persistent inflammation observed in these subjects. We will test our hypothesis by exploring caspase-1 and caspase-3 activation, as well as CD4 T-cell death by pyroptosis or apoptosis in GALT and lymph nodes from HIV-infected subjects with acute or chronic infection who are either receiving or not receiving cART (Aim 1). We will also study HIV controllers with naturally low viral loads, and immunological non-responders who control viremia on cART, but fail to recover lost CD4 T cells, likely because of persistent inflammation. We hypothesize that SAMHD1 is a two-faced host restriction factor that either prevents or promotes pyroptosis, depending on infection conditions. While it capably protects resting CD4 T cells from infection by cell-free HIV particles, when the virus is passed with high efficiency across virological synapses formed during cell-to-cell transmission, SAMHD1 turns on the host, triggering pyroptotic death pathway and chronic inflammation. If proven, these findings would radically alter our view of SAMHD1 biology (Aim 2). Finally, we hypothesize that less pathogenic forms of lentiviral infection have evolved to induce less pyroptosis and, consequently, less chronic inflammation and immune activation. We will evaluate HIV-2 infection in humans and SIVsm infection in sooty mangabeys (non- pathogenic) versus rhesus macaques (pathogenic), examining levels of caspase-1 activation and pyroptosis (Aim 3). Together, these studies promise to extend our fundamental understanding of HIV pathogenesis and the potential key role of pyroptosis as a driver of disease progression. These studies may also support the repurposing of host-directed caspase-1 inhibitors like VX-765, which is safe and well tolerated in humans, as an independent or adjunctive therapy with cART for treating HIV-infected subjects.